Application of Orthogonal Least Square (OLS) for selection of Mel frequency cepstrum coefficients for classification of spoken letters using MLP classifier

Application of Orthogonal Least Square (OLS) for selection of Mel frequency cepstrum coefficients for classification of spoken letters using MLP classifier

This paper describes an application of the Orthogonal Least Squares (OLS) algorithm for feature selection of spoken letters. Traditionally used for system identification purposes, the OLS method was used to select important Mel-Frequency Cepstrum Coefficients (MFCC) for classification of two spoken...

Full description

Bibliographic Details
Published in:Proceedings - 2011 IEEE 7th International Colloquium on Signal Processing and Its Applications, CSPA 2011
Main Author: Rozali M.F.; Yassin I.M.; Zabidi A.; Mansor W.; Tahir N.Md.
Format: Conference paper
Language:English
Published: 2011
Online Access:https://www.scopus.com/inward/record.uri?eid=2-s2.0-79957499576&doi=10.1109%2fCSPA.2011.5759923&partnerID=40&md5=7e8f6b7bd08f70db84b44939a3109031
id 2-s2.0-79957499576
spelling 2-s2.0-79957499576
Rozali M.F.; Yassin I.M.; Zabidi A.; Mansor W.; Tahir N.Md.
Application of Orthogonal Least Square (OLS) for selection of Mel frequency cepstrum coefficients for classification of spoken letters using MLP classifier
2011
Proceedings - 2011 IEEE 7th International Colloquium on Signal Processing and Its Applications, CSPA 2011


10.1109/CSPA.2011.5759923
https://www.scopus.com/inward/record.uri?eid=2-s2.0-79957499576&doi=10.1109%2fCSPA.2011.5759923&partnerID=40&md5=7e8f6b7bd08f70db84b44939a3109031
This paper describes an application of the Orthogonal Least Squares (OLS) algorithm for feature selection of spoken letters. Traditionally used for system identification purposes, the OLS method was used to select important Mel-Frequency Cepstrum Coefficients (MFCC) for classification of two spoken letters - 'A' and 'S' using Multi-Layer Perceptron (MLP) neural network. We evaluated several network structures and parameters to determine the best performance in terms of accuracy and speed. The result found that OLS is an effective feature selection method, with the best classification performance of 85% with 6 hidden units. © 2011 IEEE.


English
Conference paper
author Rozali M.F.; Yassin I.M.; Zabidi A.; Mansor W.; Tahir N.Md.
spellingShingle Rozali M.F.; Yassin I.M.; Zabidi A.; Mansor W.; Tahir N.Md.
Application of Orthogonal Least Square (OLS) for selection of Mel frequency cepstrum coefficients for classification of spoken letters using MLP classifier
author_facet Rozali M.F.; Yassin I.M.; Zabidi A.; Mansor W.; Tahir N.Md.
author_sort Rozali M.F.; Yassin I.M.; Zabidi A.; Mansor W.; Tahir N.Md.
title Application of Orthogonal Least Square (OLS) for selection of Mel frequency cepstrum coefficients for classification of spoken letters using MLP classifier
title_short Application of Orthogonal Least Square (OLS) for selection of Mel frequency cepstrum coefficients for classification of spoken letters using MLP classifier
title_full Application of Orthogonal Least Square (OLS) for selection of Mel frequency cepstrum coefficients for classification of spoken letters using MLP classifier
title_fullStr Application of Orthogonal Least Square (OLS) for selection of Mel frequency cepstrum coefficients for classification of spoken letters using MLP classifier
title_full_unstemmed Application of Orthogonal Least Square (OLS) for selection of Mel frequency cepstrum coefficients for classification of spoken letters using MLP classifier
title_sort Application of Orthogonal Least Square (OLS) for selection of Mel frequency cepstrum coefficients for classification of spoken letters using MLP classifier
publishDate 2011
container_title Proceedings - 2011 IEEE 7th International Colloquium on Signal Processing and Its Applications, CSPA 2011
container_volume
container_issue
doi_str_mv 10.1109/CSPA.2011.5759923
url https://www.scopus.com/inward/record.uri?eid=2-s2.0-79957499576&doi=10.1109%2fCSPA.2011.5759923&partnerID=40&md5=7e8f6b7bd08f70db84b44939a3109031
description This paper describes an application of the Orthogonal Least Squares (OLS) algorithm for feature selection of spoken letters. Traditionally used for system identification purposes, the OLS method was used to select important Mel-Frequency Cepstrum Coefficients (MFCC) for classification of two spoken letters - 'A' and 'S' using Multi-Layer Perceptron (MLP) neural network. We evaluated several network structures and parameters to determine the best performance in terms of accuracy and speed. The result found that OLS is an effective feature selection method, with the best classification performance of 85% with 6 hidden units. © 2011 IEEE.
publisher
issn
language English
format Conference paper
accesstype
record_format scopus
collection Scopus
_version_ 1809677914435747840

Similar Items